Intel Iris Pro Graphics 6200 vs NVIDIA GeForce GTX 765M Comparison

Head-to-Head Benchmarks

The recorded head-to-head data presents a sharply split picture. Across three synthetic workloads, each GPU claims victory in a different environment, and the margins are anything but uniform. The most dramatic result belongs to the Intel Iris Pro Graphics 6200, which absolutely dominates the NVIDIA GeForce GTX 765M in the Geekbench Metal test. The Intel part scores 7764 against the NVIDIA part's 2612, a delta of 197.2% in favor of the integrated solution. That is not a small edge; it is a nearly threefold advantage, and it suggests the Iris Pro's driver stack and architecture are unusually well optimized for Apple's Metal API, or at least for the specific workload Geekbench Metal exercises.

The reverse is true in the OpenCL test, where the NVIDIA GeForce GTX 765M reasserts its discrete-GPU pedigree. The GTX 765M scores 7176, while the Iris Pro manages only 4556. That is a 36.5% deficit for the Intel part, a decisive margin that flips the narrative. OpenCL is a general-purpose compute API, and the GTX 765M's larger shading unit count and dedicated memory subsystem clearly pay off here. The Vulkan test is closer, but still a win for NVIDIA: 6714 versus 6032, a 10.2% gap. Vulkan is a low-level, cross-platform API, and the GTX 765M edges ahead, though the margin is modest compared to the swings in Metal and OpenCL.

What does this mean in aggregate? The Intel Iris Pro takes one win out of three, but that win is staggering in magnitude. The NVIDIA part takes two wins, but its largest margin is less than a fifth of Intel's biggest. If you weight by sheer dominance in a single test, the Iris Pro's Metal result is the most notable data point in this comparison. If you weight by consistency across multiple APIs, the GTX 765M holds the edge. The average benchmark scores reflect this tension: the Iris Pro sits at 6117, the GTX 765M at 5501, a 616-point gap in Intel's favor. That average is pulled up heavily by the Metal outlier, so it should not be read as a general-purpose superiority claim.

Looking at how each GPU stacks against its own nearest rivals, the picture gets more nuanced. The Iris Pro's average of 6117 is within 0.3% of the AMD Radeon HD 8690M's 6137, and it sits 0.6% above the NVIDIA RTX A400's 6078. The GTX 765M's average of 5501 is 0.1% below the NVIDIA GeForce MX130's 5508 and 0.6% above the NVIDIA Quadro M4000's 5467. Neither GPU is a class leader; both sit in the mid-range of the database's GPU distribution, with the Iris Pro at the 35th percentile and the GTX 765M at the 32nd percentile. The data suggests these are comparable, mid-tier performers, not outliers in either direction.

FAQ

Q: Which GPU wins the most head-to-head benchmarks?

A: The NVIDIA GeForce GTX 765M wins two of the three recorded tests: Geekbench OpenCL (7176 versus 4556) and Geekbench Vulkan (6714 versus 6032). The Intel Iris Pro Graphics 6200 wins the remaining test, Geekbench Metal, with a score of 7764 versus 2612.

Q: How large is the Intel Iris Pro's win in the Metal benchmark?

A: The delta is 197.2% in favor of the Intel part. That is the single biggest margin in the entire head-to-head comparison, dwarfing the NVIDIA part's wins in OpenCL and Vulkan.

Q: Is the NVIDIA GeForce GTX 765M faster in every compute API?

A: No. It wins in OpenCL and Vulkan, but it loses decisively in Metal. The Intel Iris Pro's Metal score of 7764 is nearly triple the GTX 765M's 2612, so the NVIDIA part's advantage is not universal.

Q: How do the average benchmark scores compare?

A: The Intel Iris Pro has a higher average benchmark score at 6117, compared to the GTX 765M's 5501. That is a difference of 616 points, or roughly 11.2% relative to the NVIDIA part's average.

Q: Where does each GPU rank among all GPUs in the database?

A: The Intel Iris Pro sits at the 35th percentile, while the NVIDIA GeForce GTX 765M sits at the 32nd percentile. Both are in the lower half of the distribution, though the Intel part holds a slight percentile advantage.

Q: What are the closest rivals to each GPU?

A: The Intel Iris Pro's nearest rival is the AMD Radeon HD 8690M, with an average score of 6137, a 0.3% difference. The GTX 765M's nearest rival is the NVIDIA GeForce MX130, with an average score of 5508, a 0.1% difference.

The Verdict

The data does not support a simple "this GPU is better" conclusion. The Intel Iris Pro Graphics 6200 and NVIDIA GeForce GTX 765M are closely matched in overall average score, but they excel in entirely different API environments. If the workload is Metal-based, the Iris Pro is the clear choice; its 197.2% lead in that test is the most emphatic result in the entire comparison. No other benchmark in this head-to-head comes close to that margin. For OpenCL or Vulkan workloads, the GTX 765M is the better pick, with leads of 36.5% and 10.2% respectively. Those are not trivial advantages, especially the OpenCL one.

Who should pick which? A user whose primary applications run on Apple's Metal API, or who values a single, decisive performance outlier, would lean toward the Iris Pro. It is also the lower-power option by a wide margin, a factor that matters in thin-and-light systems. A user running OpenCL compute tasks, or Vulkan-based games or applications, would prefer the GTX 765M. It delivers more consistent wins across two of the three tested APIs, and its memory subsystem is dedicated rather than system-shared, which the benchmark results reflect. Neither GPU is a top-tier part; both sit near the 32nd to 35th percentile of all GPUs. The choice is workload-dependent, not absolute.

Specification Differences

The two GPUs diverge on nearly every hardware specification. The Intel Iris Pro is built on a 14 nm process, while the NVIDIA GeForce GTX 765M uses a 28 nm process. That is a significant generational gap in manufacturing technology. The Iris Pro's base clock is 300 MHz with a boost of 1100 MHz; the GTX 765M's base clock is 797 MHz with a boost of 863 MHz. The NVIDIA part starts higher but boosts less, while the Intel part has a much wider boost range.

Memory is another major difference. The Iris Pro uses system-shared memory, with bandwidth described as "System Dependent." The GTX 765M has 2 GB of dedicated GDDR5 memory on a 128-bit bus, delivering 64.13 GB/s of bandwidth. That dedicated memory is almost certainly a factor in the GTX 765M's OpenCL and Vulkan wins. The shading unit count also favors NVIDIA: 768 shading units on the GTX 765M versus 384 on the Iris Pro. Texture mapping units are 64 versus 48, and ROPs are 16 versus 6. The pixel rate is 13.81 GPixel/s for NVIDIA versus 6.600 GPixel/s for Intel, a 2.1x advantage. Texture rate is closer: 55.23 GTexel/s versus 52.80 GTexel/s. FP32 compute is 1,325.6 GFLOPS for the GTX 765M versus 844.8 GFLOPS for the Iris Pro.

Power consumption differs dramatically. The Iris Pro is rated at 15 W, while the GTX 765M is rated at 75 W. That is a fivefold gap. The Iris Pro is an IGP, meaning it is integrated into the CPU package, while the GTX 765M is an MXM Module with a slot width of its own. The bus interface is Ring Bus for Intel versus MXM-B (3.0) for NVIDIA. Display outputs are motherboard-dependent for the Intel part and portable-device-dependent for the NVIDIA part. The APIs also differ: the Iris Pro supports DirectX 12 (11_1), OpenGL 4.4, and Vulkan 1.0. The GTX 765M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The NVIDIA part has a higher OpenGL and Vulkan version, which may explain part of its Vulkan benchmark lead.

Architecture Differences

The architectural story is one of two very different design philosophies. The Intel Iris Pro Graphics 6200 is built on the Broadwell GT3e chip, using Intel's Generation 8.0 architecture. It is part of the HD Graphics (Broadwell) generation. The NVIDIA GeForce GTX 765M is built on the GK106 chip, using NVIDIA's Kepler architecture, and belongs to the GeForce 700M generation. Intel's process node is 14 nm, produced at Intel's own foundry. NVIDIA's process node is 28 nm, produced at TSMC. The GTX 765M's transistor count is listed at 2,540 million, with a die size of 221 mm² and a transistor density of 11.5M per mm². The Iris Pro has no transistor or die size data recorded, but given the 14 nm node and integrated nature, the comparison is lopsided in density.

The cache and core structures differ as well. The Iris Pro has 384 shading units, 48 TMUs, and 6 ROPs. The GTX 765M has 768 shading units, 64 TMUs, and 16 ROPs. That is a 2x advantage in shading units and ROPs, and a 1.33x advantage in TMUs for NVIDIA. The Iris Pro's memory architecture relies entirely on the system's shared memory, with no dedicated VRAM. The GTX 765M has its own 2 GB GDDR5 pool. The Iris Pro's memory clock is listed as "System Shared," while the GTX 765M's memory runs at 1002 MHz, or 4 Gbps effective. The Iris Pro's bandwidth is "System Dependent," a phrase that acknowledges the variability of shared-memory performance. The GTX 765M's bandwidth is a fixed 64.13 GB/s.

The release dates are close but not identical. The Iris Pro was released on 2014-09-04, while the GTX 765M was released on 2013-05-29. Both are marked as end-of-life production status. The GTX 765M has a predecessor (GeForce 600M) and a successor (GeForce 800M), while the Iris Pro has neither recorded. The power envelope is the starkest architectural difference: 15 W for the integrated Intel part versus 75 W for the discrete NVIDIA module. That 60 W gap explains why the Iris Pro can be integrated into a CPU die, while the GTX 765M requires a separate MXM module with its own cooling and power delivery.

Where Each One Wins

The Intel Iris Pro Graphics 6200 wins in the Metal benchmark by a massive 197.2% margin. That is its one clear victory, and it is a dominant one. The data implies that this GPU is exceptionally well-suited to Metal-based workloads, likely due to driver-level optimizations or architectural features that align with Metal's low-level design. It also wins on power efficiency: 15 W versus 75 W is a fivefold difference, and for any system where thermal or battery constraints matter, the Iris Pro is the only rational choice. Its average benchmark score is also higher, at 6117 versus 5501, so in a pure "average across all tested workloads" sense, it comes out ahead.

The NVIDIA GeForce GTX 765M wins in OpenCL by 36.5% and in Vulkan by 10.2%. Those are the two compute-heavy, cross-platform APIs in the test, and the GTX 765M's dedicated 2 GB GDDR5 memory and 768 shading units give it a real advantage in those environments. It also wins on raw hardware specs: higher clock base, more shading units, more TMUs, more ROPs, higher pixel rate, higher texture rate, and nearly 1.6x the FP32 compute throughput. For any workload that scales with shading unit count or memory bandwidth, the GTX 765M is the better bet. Its OpenGL 4.6 support also exceeds the Iris Pro's OpenGL 4.4, which could matter for older or OpenGL-specific applications.

The split is clean: Intel wins the Apple-centric Metal test and the efficiency battle, NVIDIA wins the general-purpose compute tests and the raw hardware specification comparison. A user building a lightweight, low-power system with Metal-based software should choose the Iris Pro. A user running OpenCL or Vulkan workloads on a machine that can tolerate a 75 W discrete module should choose the GTX 765M. The data does not crown a single winner; it describes two GPUs with complementary strengths.